COP 3330 Exam 2 Review
Chapter Sections
7
7.1 – 7.8
Topics
Classes, static vs. instance,
this, references vs. objects,
inherited methods, overriding,
scoping rules, method overloading,
generic classes
None
None
Aggregation
8
8.3 – 8.4,
8.8 – 8.11
Iterator, Array Processing,
2-D Arrays, Vector, Arraylist
Collection Algorithms
9
9.1 – 9.2
Inheritance,
Polymorphism,
Visibility Modifiers
How to study:
1) Go through both of your programs. Make sure you
understand how they work. I may ask questions based on your
homework assignments.
2) Look over all of the sample program presented in class. For
this semester, these contain a more thorough breakdown of the
material covered in class than the notes.
2) Read the class notes. Make sure you understand the ideas
presented.
3) Read the sections denoted above.
4) Peruse the problems at the end of the appropriate chapters
in the text and plan out how you would attack them.
Exam Format
Here are the different types of questions I will have:
1) Tracing: Either through code segments or of an entire
program that uses a class.
2) Writing: You may have to write a main method that uses a
class, or write class methods.
3) You may have to find mistakes in a piece of code.
4) Short Answer: Some of the reading goes over general
concepts in fairly non-technical manner. I may ask questions
on this material that require a single sentence answer.
5) Design Question: I may give you the basic idea of a class to
be written and then ask you general questions about the design
of the class. (What methods should be included? What should
be private, public, etc.?)
Exam Aids, etc.
The exam will be open book and WILL NOT consist of a
multiple choice section like the previous exam. Make sure to
bring your text. Notes will not be allowed.
Defining Classes
Remember when you are defining a class, you need to define
two parts:
1) instance variables (these dictate the structure of an object
of the class.)
2) instance methods (these dictate what operations are allowed
on an object of the class.)
Generally instance variables are private, and all methods you
want others to use are public. Other methods, such as the
totalminutes method in the Time class need not be public.
Remember that there are three types of variables you could
have when dealing with instance methods:
1) instance variables
2) formal parameters
3) local variables
Learn the difference between the three and do not call
different types of variables with the same name.
Remember that anytime you refer to an instance variable
in an instance method, you are referring to the instance
variable OF the object the method was called upon.
Formal parameters are the information the method needs to
complete it's task.
Local variables are declared as necessary, such as loop indexes.
Constructors
The job of a constructor is to initialize an object of the class.
Unlike C where you have to specifically allocate space for the
object with a malloc call, (or something similar), java takes
care of finding the appropriate memory automatically. All you
have to do is initialize the components of the object.
All constructors are public and must be named the same thing
as the class. A constructor in the Time class has the signature:
public Time(int h, int m);
One important thing to keep straight in a constructor is the
order of the assignment statements. In particular, the instance
variables need to be assigned, so these need to be on the lefthand side and the formal parameters need to be on the right.
Constructors are often overloaded, meaning that there are
multiple ones defined with different parameter lists. The one
instantiated depends on the number and types of the actual
parameters. A default constructor is one that takes in no
parameters.
Inheriting Methods
All classes in java are part of an inheritance structure.
In particular, all classes inherit from the class Object. From
there, we can have a longer “chain” of inheritance.
In any inheritance relationship, the class from which another
class inherits is known as the base/parent class. The class that
does the inheriting is known as the sub/child class. The child
class retains all characteristics of the parent class, plus can
have some unique characteristics of its own.
Thus, the Time class, which we defined in class, automatically
inherits all of the characteristics of the Object class. For
example, the Object class has a toString method. If there is no
toString method declared in the Time class, then anytime the
toString method is called on a Time object, it will actually run
the toString method defined in the Object class. In essence, the
Time class has inherited the toString method in this example.
The other option would be to write a toString method in the
Time class. Once we do this, if we call the toString method on a
Time object, it will call the one defined in the Time class,
overriding the toString method in the Object class. (The rule is
that the method in the “nearest” class overrides all similar
methods in previous parent classes.)
It would be a good idea to take some of the examples shown in
class and test out what occurs when a method is overridden
and when it is not.
toString
The following method is defined in the Object class:
public String toString();
It is frequently overridden in user-defined classes. The reason
for this is as follows:
Anytime someone tries to print out an Object, what java
automatically does is call toString() on that Object and just
prints out what that method returns. Thus, if t is a time object
and you run
System.out.println(t); //You are implicitly running
System.out.println(t.toString());
compareTo
This method is necessary to implement the Comparable
interface. We will talk about interfaces later, but the basic idea
is that to implement an interface, you must write certain
methods in your class. To implement Comparable, you must
write a method with the following signature:
public int compareTo(Object o);
Furthermore, the method should work as follows:
It should return a negative integer if the current object comes
before o. It should return 0 if the two objects are equal and it
should return a positive integer if the current object comes
after o.
Notice that the method takes in an Object. Thus, you must cast
o to the appropriate type in order to access any components of
the actual object that gets passed into the method.
static vs. instance
static means, “belongs to the class.”
instance means, “belongs to the object.”
An instance method MUST BE called on an object of that
class.
A
static
method
is
always
called
CLASSNAME.method(<parameter list>); because it just
belongs to the class and does NOT operate on any object
inherently.
An instance variable belongs to the object, whereas only one
copy of a static variable ever exists, regardless of how many
objects of a class are instantiated.
Visibility modifiers
For right now, we are only using two of them:
Private: Accessible only inside of the class.
Protected: Accessible inside the class, subclass and package.
Public: Accessible anywhere.
Usually, instance variables are private while most instance
methods are public. The goal here is to hide the representation
of the data from the user. The public methods are the only
means by which the user can manipulate objects they create.
We can restrict the types of operations the user can perform by
restricting the public methods we give them access to.
Aggregation
When we design a set of classes for an application, one of the
relationships between classes is aggregation. This relationship
is when one class contains OR HAS-A object of another class.
For example, a Lemonade Object is comprised of several
components, one of which is a BankAccount object. In many
instances it makes sense to create classes with an aggregation
relationship. (If you created a Car object, it would ideally
contain/have several wheel objects.)
It's important to recognize and understand the difference
between an aggregation relationship and an inheritance
relationship.
The best way to internalize these concepts is to look at
examples with several classes and consider how you would
extend them.
Iterator
If you want to "go through" each element in a Java collection
or array, you can do so as follows:
for (Object o : collectionName) {
// Use o in here
}
The first item inside of the loop is the type of item being stored
in the collection. The second item is the variable name
temporarily given for an arbitrary object in the collection. This
is followed by a colon and then the name of the collection
through which you want to iterate. Inside the loop, instead of
having to index into an array, just use the given variable name
in the body of the loop and those actions will be performed on
each element in the collection as the loop executes.
2-Dimensional Arrays
Technically, a 2 dimensional array is an array of arrays. It is
possible to allocate space for the array references, without
allocating space for each array. Furthermore, it is possible to
allocate space for references to objects without allocating space
for each object. Once the array is created, it is indexed just like
a 2 dimensional array in C.
Object[][] vals; // vals is a reference to a 2D array
vals = new Object[10][]; // we've allocated vals to point to an
//array that stores 10 array references.
for (int i=0; i<vals.length; i++)
vals[i] = new Object[i+1]; // allocated space for each array
// individually;
for (int i=0; i<vals.length; i++)
for (int j=0; j<vals[i].length; j++)
vals[i][j] = new Object(); // create each individual Object.
Vector, ArrayList
Java provides these classes that manage collections. Generally
they are quite handy. They allow you to add elements, search
for elements and remove elements from a collection. It's
important to remember some syntactic issues when using these.
If you create a Vector of Object for example, sometimes you
may need to cast the value returned from a method to the
proper type. No casting is necessary to add an Object into a
collection.
Remember to use the Generic construction when using these
classes:
ArrayList<String> s = new ArrayList<String>();
Collections Algorithms
Java provides some very nice methods to help manipulate
collections. Here are some of the algorithms provided:
Sorting, Max, Min, Reverse, and Shuffle
All of the methods are static, thus, they are called from the
class and the collection is always passed in as a parameter.
In order to sort, the type of objects stored in the collection
must implement Comparable.
Basics of Inheritance
An inherited class is defined as follows, using the key word
extends:
public class ThreeDimensionalPoint extends
Point {
…
}
In the example above, Point is known as the base/parent/super
class, and ThreeDimensionalPoint is Point's subclass.
(Alternatively, Point is known as the base class from which
ThreeDimensionalPoint inherits. Also, there is an IS-A
relationship here – a ThreeDimensionalPoint object IS A Point
object as well.)
A subclass inherits all the characteristics of the superclass.
Thus, all public methods in the superclass are valid to be run
on objects of the subclass.
In the typical case, the subclass adds at least one instance
variable, separate from those it inherited from the superclass.
In the example in the text, ThreeDimensionalPoint has its own
instance variable z, and inherits x and y from Point.
In the typical case, a subclass just uses some of the methods in
its superclass, but overrides other methods. To override a
method, you must define a method with the same signature as
one in the superclass in the subclass. The purpose of this is so
that if you have a ThreeDimensionalPoint object, you end up
running a different method on it than if you just had a Point
object.
The method call super()
When overriding a method in the superclass, you have the
option of calling the corresponding method from the
superclass. To do this, instead of using "this" just use the name
"super". For all methods except for constructor, you would
call, super.methodname(). Pass the parameters you need/want
to pass to the corresponding method in the superclass.
Constructors
All constructors in the subclass will automatically call the
default constructor in the base class if there is NO EXPLICIT
call to the base class constructor.
The other option is to make an explicit call to the superclass
constructor using "super". In this case, you control which of
the constructors in the superclass gets called. (Here, the
method call is just super, not super.methodname())
The rationale behind this is that you MUST build the base
object first before building any of the rest of it to create the
subclass specialization.
Reasons to use Inheritance, instead of just adding to one class
Putting all the functionality into one class may make it difficult
to understand, or more difficult to use for simple things. (For
example, sometimes you want to use points in 2 dimensions,
sometimes in three. By incorporating inheritance, you make
the use of both seamless. If you didn't, when wanting a 2-D
point, you'd have to always just set z=0 and deal with that
extra component.)
Polymorphism
Definition: A code expression can invoke different methods
depending on the type of objects using the code.
Thus, polymorphism allows the same exact method call to
invoke different methods. Here's the key:
In the code, we always use references.
But, those references can refer to any object of the class of the
reference, OR any SUBCLASS (direct or indirect) of the class
of the reference.
So, for example, we have a ColoredPoint class that inherits
from Point, and a ThreeDColoredPoint class that inherits from
ColoredPoint, a Point reference could refer to a Point object, a
ColoredPoint object, or a ThreeDColoredPoint object.
Java dynamically determines the actual exact class of the
object and uses that information when deciding what method
to invoke, when there are options from multiple classes.
Rule 1: First look for the appropriate method IN the actual
exact class of the object. If it's not there, then look in the direct
superclass. If it's not there, continue up the inheritance
structure.
Rule 2: For a match to occur, the name, number of
parameters, and types of corresponding parameters must
match.
Remember, the type match is NOT polymorphic. Thus, if a
parameter is a reference from a base class, but that reference is
referring to an object of the subclass, and that subclass is the
parameter type, then this will NOT be considered a match.
Visibility Modifier Rules
Member
Restriction
public
protected
default
private
this
Subclass
Package
General
X
X
X
X
X
X
X
X
X
X
instanceof operator
The instanceof operator is used in Boolean expressions as
follows:
reference instanceof Class
This expression is true if the reference to the left of the
operator is referring to an object that belongs in the class
designated to the right of the operator. It doesn't have to be the
exact class. Thus, if the variable pt is a Point reference, and it
was referring to a ThreeDColoredPoint object, then the
expressions
pt instanceof Point
pt instanceof ColoredPoint
pt instanceof ThreeDColoredPoint
would all evaluate to true.
But if pt referred to a ColoredPoint object, then the first two
would be true and the last one would be false.